Testing for Bubble(s) in NASDAQ and DJI indexes in 1990-2003
Author
Fayazbakhsh, Seyed Kian
Term
4. semester
Publication year
2022
Submitted on
2022-12-14
Pages
69
Abstract
This thesis examines whether the Nasdaq Composite index experienced a price bubble between 1990 and 2003—a period when prices may rise unsustainably fast. To detect “explosive” price behavior, it uses two main methods. The GSADF test identifies windows of rapid, self-reinforcing growth and can date-stamp those episodes. The variance bounds test assesses whether price volatility exceeds levels consistent with fundamentals. To better understand the timing and nature of any explosive periods, the study also applies supporting diagnostics: the Chow break test (to detect structural breaks), an abnormal return test (to flag returns that deviate from expected patterns), and a variance ratio test (to evaluate whether prices follow a random walk). These additional tests aid interpretation but are not designed to confirm bubbles on their own.
Denne afhandling undersøger, om Nasdaq Composite-indekset oplevede en prisboble mellem 1990 og 2003—en periode hvor kurser kan stige uholdbart hurtigt. For at finde eksplosiv prisadfærd anvendes to hovedmetoder. GSADF-testen kan tidsfeste perioder med hurtig, selvforstærkende vækst og indikere, om en boble har været til stede. Variance bounds-testen vurderer, om udsving i priserne overstiger niveauer, der er forenelige med fundamentale forhold. For bedre at forstå timingen og karakteren af eventuel eksplosivitet suppleres analysen med diagnostiske tests: Chow break-testen (finder strukturelle brud), en abnormal return-test (peger på afkast, der afviger fra forventede mønstre), og en variance ratio-test (vurderer om priserne følger en tilfældig gang/“random walk”). Disse ekstra tests hjælper med fortolkningen, men er ikke i sig selv designet til at bekræfte bobler.
[This apstract has been rewritten with the help of AI based on the project's original abstract]
Keywords